Error in Merging Variant Callsets from Multiple Callers into one VCF File

[lkuang]

Here is my cmd used:

 pbrun votebasedvcfmerger \
		    --in-vcf strelka:HG002.hiseqx.pcr-free.30x.pb.strelka.strelka_work/results/variants/variants.vcf.gz \
		    --in-vcf DeepVariant:HG002.hiseqx.pcr-free.30x.pb.deepvariant.vcf \
		    --in-vcf HaplotypeCaller:HG002.hiseqx.pcr-free.30x.pb.haplotypecaller.vcf \
		    --min-votes 1 \
		    --out-dir caller_intersection.vcf

Output is like this:

Then I use bgzip to get HG002.hiseqx.pcr-free.30x.pb.deepvariant.vcf.zp and HG002.hiseqx.pcr-free.30x.pb.haplotypecaller.vcf.gz and run

pbrun votebasedvcfmerger \
				    --in-vcf strelka:HG002.hiseqx.pcr-free.30x.pb.strelka.strelka_work/results/variants/variants.vcf.gz \
				    --in-vcf DeepVariant:HG002.hiseqx.pcr-free.30x.pb.deepvariant.vcf.gz \
				    --in-vcf HaplotypeCaller:HG002.hiseqx.pcr-free.30x.pb.haplotypecaller.vcf.gz \
				    --min-votes 1 \
				    --out-dir caller_intersection.vcf

Errors become this:

[gburnett]

Hey @lkuang,

Have it looks like your vcf files still need to be indexed. For this I would recommend tabix.

Did these VCFs come from Parabricks, or did you run them with another tool?

[lkuang]

Hi @gburnett , I do that following the instructions in the pb3.7 manual. The VCF files are generated from pb3.7.

I will try tabix.

[lkuang]

Thanks @gburnett. I just understand that the bgzip and tabix are both necessary. So it’s OK now.

BTW, I think it will be good if these steps cmd or instructions could be added in the manual as it is an step-by-step how-to.